Query Answering with Objective-Subjective Retrieval Routing
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional systems are limited in providing both objective and subjective recommendations in response to user queries, failing to address subjective queries such as brand recommendations or tips on how to use a product effectively.
Innovation Solution
A system and method that utilizes machine learning and natural language processing to classify queries as objective or subjective, and retrieves appropriate answers from databases, including online forums and product catalogs, to provide real-time responses that include both objective information and subjective recommendations.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If conventional FAQ-based approaches are used to answer user queries, then objective information can be provided efficiently, but subjective recommendations cannot be provided
Solution Approach 1:
The system segments the answer retrieval process into two distinct pathways: an objective answer retrieval module for factual information and a subjective answer retrieval module for recommendations and opinions. This segmentation allows the system to handle different query types appropriately without requiring a complete system redesign, thus expanding versatility while managing complexity.
Solution Approach 2:
The system creates a universal query processing framework that can handle both objective and subjective queries through a single interface. The query processing module automatically determines the query type and routes to the appropriate retrieval module, making the system multi-functional without requiring users to interact with separate systems.
2Adaptability or versatility
If the system retrieves answers from multiple data sources including online forums, then subjective recommendations can be provided, but response time may increase
Solution Approach 1:
The system performs preliminary classification of the query type (objective or subjective) immediately upon receiving the query, before initiating the retrieval process. This preliminary action allows the system to select the appropriate retrieval module in advance, avoiding unnecessary processing steps and reducing response time while maintaining the capability to access multiple data sources.
Solution Approach 2:
The system extracts and processes only the relevant portion of data from multiple sources based on the query type. For subjective queries, it specifically extracts recommendations and opinions from online forums rather than processing all available data, thus reducing processing time while maintaining the capability to provide subjective recommendations.
3Measurement precision
If the system uses machine learning to classify queries in real-time, then query accuracy improves, but processing complexity increases
Solution Approach 1:
The system introduces a query processing module as an intermediary between the user query and the retrieval modules. This intermediary contains the machine learning classification logic and acts as a mediator that translates complex query analysis into simple routing decisions, improving classification accuracy while shielding the rest of the system from processing complexity.
Data Source
AI summary
A method including determining respective training data for each of query-type-specific answer retrieval modules. The method further can include training each of the query-type-specific answer retrieval modules. The method additionally can include determining a query type of a query from a user device for a user. The method also can include determining, in real-time and based at least in part on the query type, an answer retrieval module from the query-type-specific answer retrieval modules, as trained. Moreover, the method can include determining, in real-time by the answer retrieval module, one or more answers for the query. Then, the method can include ranking, in real-time, the one or more answers based on a user profile of the user. Finally, the method can include transmitting, via a computer network and to the user device, at least one of the one or more answers, as ranked. Other embodiments are disclosed.